Skip to contentSkip to conversation

More UDMA Troubles

DISCUSSION Andre LeBlanc 5 participants 7 messages May 13, 2002

lore
Share & tools
lore
#0 lore raw

Ok, I signed up to this list from work under the name aeleblanc, now i'm at home, so to see this history look at the other messages, anyway, i was having alot of troubles getting DMA Working on my system, its a duron 1GHz on an ECS Motherboard w/ an SiS Chipset. Actually, even with DMA Disabled i was getting Hard drive errors... anyway, I tried compiling 2.5.15 because i was told that 2.4.19-pre8 and up had better IDE Support, but during the boot process on the new Kernel I get the following messages

hda: dma_intr: error=0x84 {DriveStatusError BadCRC }
hda: recalibrating!
{ dma_intr }
hda: dma_intr: error=0x84 {DriveStatusError BadCRC } { dma_intr }
hdb: DMA Disabled

then the system Locks solid.

can anyone help me with this, this is the third kernel I've tried and they've all had very serisous problems.

#1 lore raw

Sorry!!!

Sorry for the multiple posts... I waited a long time and they weren't showing up.

#2 lore raw
In reply to Andre LeBlanc

Hi Andre <deja vu feeling>

Did this happen in 2.4? If so please give me the exact kernel because doing DMA on a PIO transaction is a driver bug!

Quoted 21 lines · Andre LeBlanc

Ok, I signed up to this list from work under the name aeleblanc, now i'm at home, so to see this history look at the other messages, anyway, i was having alot of troubles getting DMA Working on my system, its a duron 1GHz on an ECS Motherboard w/ an SiS Chipset. Actually, even with DMA Disabled i was getting Hard drive errors... anyway, I tried compiling 2.5.15 because i was told that 2.4.19-pre8 and up had better IDE Support, but during the boot process on the new Kernel I get the following messages

hda: dma_intr: error=0x84 {DriveStatusError BadCRC }
hda: recalibrating!
{ dma_intr }
hda: dma_intr: error=0x84 {DriveStatusError BadCRC } { dma_intr }
hdb: DMA Disabled

then the system Locks solid.

can anyone help me with this, this is the third kernel I've tried and they've all had very serisous problems.

Quoted 6 lines

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Andre Hedrick
LAD Storage Consulting Group

#3 lore raw
In reply to Andre LeBlanc

Andre: This occured on 2.5.15 but I also had similar errors with 2.4.18, and the default kernel thats installed with debian woody (I think its 2.2.17, not sure though), but the system still booted, it just did strange things afterwards and eventually locked up. the 2.2.17 worked fine until I did hdparm -c3 -d1 -X34.

Lionel: I don't want to sound like an idiot or anything but I have no idea how to install 2.4.19-pre8, it looks like some kind of patch, but i've never used one before.

I think that running that, hdparm command may have actually done some damage to my computer... since then my BIOS Occasionally doesn't recognize my hard drive, and Windows 2000 keeps getting bluescreens during the boot process, it took 4 tries to get it to boot properly, and I ahve a feeling it will lock up eventually. (I've never had a bluescreen uder 2000 before.)

Quoted 1 line · Andre LeBlanc
Quoted 1 line

Ok, I signed up to this list from work under the name aeleblanc, now i'm

at

Quoted 2 lines
Quoted 2 lines

home, so to see this history look at the other messages, anyway, i was having alot of troubles getting DMA Working on my system, its a duron

1GHz

Quoted 1 line
Quoted 1 line

on an ECS Motherboard w/ an SiS Chipset. Actually, even with DMA

Disabled i

Quoted 1 line
Quoted 1 line

was getting Hard drive errors... anyway, I tried compiling 2.5.15

because i

Quoted 1 line
Quoted 1 line

was told that 2.4.19-pre8 and up had better IDE Support, but during the

boot

Quoted 5 lines
Quoted 1 line

process on the new Kernel I get the following messages

Please try 2.4.19-pre8. 2.5 SiS IDE driver is not yet updated.

LB.

#4 lore raw
Quoted 5 lines

I think that running that, hdparm command may have actually done some damage to my computer... since then my BIOS Occasionally doesn't recognize my hard drive, and Windows 2000 keeps getting bluescreens during the boot process, it took 4 tries to get it to boot properly, and I ahve a feeling it will lock up eventually. (I've never had a bluescreen uder 2000 before.)

hdparm just sets configuration variables, they should be lost on physical power off but might survive a warm boot (even then the bios ought to have shoved things back into its idea of ide happy mode)

#5 lore raw
In reply to Andre LeBlanc
Quoted 10 lines · Andre LeBlanc

Andre: This occured on 2.5.15 but I also had similar errors with 2.4.18, and the default kernel thats installed with debian woody (I think its 2.2.17, not sure though), but the system still booted, it just did strange things afterwards and eventually locked up. the 2.2.17 worked fine until I did hdparm -c3 -d1 -X34.

Lionel: I don't want to sound like an idiot or anything but I have no idea how to install 2.4.19-pre8, it looks like some kind of patch, but i've never used one before.

From memory :

go to http://www.kernel.org/
1/ get linux-2.4.18.tar.gz as you probably did for 2.5.15.
2/ before configuring the kernel sources:
- get patch-2.4.19-pre8.gz,
- apply the patch by:
. going into the 2.4.18 source directory,
. running zcat <path_to_download_dir>/patch-2.4.19-pre8.gz | patch -p1
3/ configure, compile, install.

Quoted 8 lines

I think that running that, hdparm command may have actually done some damage to my computer... since then my BIOS Occasionally doesn't recognize my hard drive, and Windows 2000 keeps getting bluescreens during the boot process, it took 4 tries to get it to boot properly, and I ahve a feeling it will lock up eventually. (I've never had a bluescreen uder 2000 before.)

Strange.
I've a similar problem here with an ECS K7S5AL where one of the 4 IDE devices seems to transmit bad data on a "cold" boot, this drive's id is usually "Maxtor 4G120J6" but on a cold boot the BIOS see everytime a L instead of a M in "Maxtor" and other characters are shifted (--ascii_value or maybe low bit of ascii_value zeroed, didn't check) at regular intervals. The whole data on the disk seems to be affected as Linux can't mount the only fs on the drive. I moved the disk from hdc to hdd and then to hdb and each time it behaves like described above. Updating the BIOS doesn't solve the problem. The only workaround is to wait several minutes for the system to warm up.
After warm up I *never* encoutered a crash with 2.4.19-pre7-ac2 which holds the very same SiS IDE driver than 2.4.19-pre8 (at least 14 days uptime doing regular video grabbing on another disk and occasionnal huge file transfers between the Maxtor and other disks).

In my case the problem must be the drive firmware or the hardware itself. This may be the same problem for you. Do you have another drive to make tests with ? When your BIOS doesn't recognise your drive, what's the error ? When the BIOS recognise it and Windows bluescreens, what's the drive ID reported by the BIOS ?

You didn't tell us what your mainboard model is, could you please do so and tell us the harddrive model (hdparm -i) ?

LB.

#6 lore raw
In reply to Andre LeBlanc
Quoted 3 lines · Andre LeBlanc

I think that running that, hdparm command may have actually done some damage to my computer...

Highly unlikely -

Quoted 5 lines

since then my BIOS Occasionally doesn't recognize my hard drive, and Windows 2000 keeps getting bluescreens during the boot process, it took 4 tries to get it to boot properly, and I ahve a feeling it will lock up eventually. (I've never had a bluescreen uder 2000 before.)

I've seen win 2000 reboot itself when a printer is turned on - and win2k using friends have reported that it reboots when starting aol instant messenger -

Joe